Install
openclaw skills install memphis๐ฅ Memphis - Complete AI Brain for OpenClaw Agents ALL-IN-ONE meta-package with everything you need: ๐ง Core Features: - Local-first memory chains (journal, recall, ask, decisions) - Offline LLM integration (Ollama, local models) - Semantic search with embeddings - Knowledge graph - Encrypted vault for secrets ๐ Cognitive Engine (Models A+B+C): - Model A: Record conscious decisions (manual) - Model B: Detect decisions from git (automatic) - Model C: Predict decisions before you make them (predictive) - 90.7% accuracy, proactive suggestions ๐ ๏ธ Setup & Management: - Bootstrap wizard (5-minute setup) - Self-loop capability (Memphis uses itself) - Auto-repair system - Chain monitoring - Backup automation ๐ Multi-Agent Network: - Campfire Circle Protocol - Share chain sync (IPFS) - Multi-agent collaboration - Agent negotiation (trade protocol) Perfect for: Individual developers, teams, researchers, entrepreneurs Quick start: clawhub install memphis && memphis init
openclaw skills install memphisTransform any OpenClaw agent into a fully-functional cognitive partner in 5 minutes!
Memphis = Agent + LLM + Memory Chain + DECIDE + PREDICT
A local-first AI brain with persistent memory chains, offline LLM integration, and complete cognitive capabilities (conscious, inferred, and predictive decisions).
๐ฆ Memphis v3.0.0
โโโ ๐ง Core Brain (v3.0.0)
โ โโโ Journal (capture memories)
โ โโโ Recall (semantic search)
โ โโโ Ask (LLM-powered Q&A)
โ โโโ Decisions (decision tracking)
โ โโโ Vault (encrypted secrets)
โ โโโ Graph (knowledge graph)
โ โโโ Embeddings (vector search)
โ
โโโ ๐ Cognitive Engine (v3.0.0)
โ โโโ Model A: Record decisions (manual)
โ โโโ Model B: Detect decisions (git-based)
โ โโโ Model C: Predict decisions (AI-powered)
โ โโโ Pattern learning (90.7% accuracy)
โ โโโ Proactive suggestions
โ
โโโ ๐ ๏ธ Setup & Management (v2.0.0)
โ โโโ Bootstrap wizard (5-minute setup)
โ โโโ Self-loop capability
โ โโโ Auto-repair system
โ โโโ Chain monitoring
โ โโโ Backup automation
โ
โโโ ๐ Multi-Agent Network (v1.0.0)
โโโ Campfire Circle Protocol
โโโ Share chain sync (IPFS)
โโโ Multi-agent collaboration
โโโ Agent negotiation
clawhub install memphis
# Interactive setup wizard
memphis init
# Or quick setup with identity
memphis init --identity "YourAgent" --role "Assistant" --location "localhost"
# Create your first memory
memphis journal "Hello Memphis! I'm your first memory!" --tags first,hello
# Search your memory
memphis ask "What is my first memory?" --provider ollama
# Record a conscious decision
memphis decide "Framework choice" "React" \
--options "React|Vue|Angular" \
--reasoning "Best ecosystem and community support" \
--tags tech,frontend
# See predicted decisions based on patterns
memphis predict
# Enable learning mode
memphis predict --learn
โ Done! Memphis is ready!
# Journal - Capture everything
memphis journal "Learned: TypeScript generics" --tags learning,typescript
memphis journal "Met with team about Project X" --tags meeting,project-x
# Recall - Semantic search
memphis recall "TypeScript" --top 20
# Ask - LLM-powered Q&A
memphis ask "What did I learn about TypeScript?" --provider ollama
# Record decision
memphis decide "Database" "PostgreSQL" \
--options "PostgreSQL|MongoDB|SQLite" \
--reasoning "Need ACID transactions" \
--tags architecture,database
# View decision
memphis show decision 1
# List all decisions
memphis decisions
# Build graph from chains
memphis graph build
# Explore connections
memphis graph show --chain journal --limit 10
# Initialize vault
memphis vault init --password-env MEMPHIS_VAULT_PASSWORD
# Add secret
memphis vault add openai-key sk-xxx --password-env MEMPHIS_VAULT_PASSWORD
# Get secret
memphis vault get openai-key --password-env MEMPHIS_VAULT_PASSWORD
# Manual decision recording
memphis decide "Use TypeScript" "TypeScript" -r "Better type safety"
# Auto-detect from git commits
memphis git-analyze --auto-decide
# Review suggestions
memphis review --pending
# Predict next decisions
memphis predict
# Enable learning
memphis predict --learn
# View patterns
memphis patterns show
# Verify chain integrity
memphis verify
# Auto-repair issues
memphis repair --auto
# Health check
memphis doctor
# Chain status
memphis status
# Create backup
memphis backup create ~/backups/memphis-$(date +%Y%m%d).tar.gz
# Restore
memphis backup restore ~/backups/memphis-20260303.tar.gz
# Setup multi-agent network
memphis network setup --partner "Memphis at 10.0.0.80"
# Sync share chain
memphis share-sync
# Send message to partner
memphis share "Working on feature X" --type update
memphis init # Initialize Memphis brain
memphis status # Health check
memphis doctor # Diagnostic
memphis journal # Add memory
memphis recall # Search memory
memphis ask # Ask LLM + memory
memphis decide # Record decision
memphis show # Show block/decision
memphis embed # Generate embeddings
memphis verify # Chain integrity
memphis repair # Fix issues
memphis backup # Backup/restore
memphis predict # Predict decisions (Model C)
memphis patterns # Pattern analysis
memphis git-analyze # Git integration (Model B)
memphis suggest # Proactive suggestions
memphis reflect # Reflection engine
memphis network # Network management
memphis share-sync # Sync with partners
memphis share # Send message
memphis trade # Agent negotiation
memphis graph # Knowledge graph
memphis vault # Encrypted secrets
memphis ingest # Import documents
memphis offline # Offline mode
memphis mcp # MCP server
memphis daemon # Background agent
# Morning routine
memphis status
memphis reflect --daily
memphis journal "Session start: Project X" --tags session
# During work
memphis decide "API design" "REST" -r "Simpler than GraphQL"
memphis journal "Learned: rate limiting" --tags learning
# End of day
memphis embed --chain journal
memphis reflect --daily --save
# Share decisions
memphis decide "Stack" "TypeScript + React" -r "Team expertise" --tags team
memphis share-sync
# Multi-agent sync
memphis network status
memphis share "Decision: Use PostgreSQL" --type decision
# Ingest papers
memphis ingest ./papers --chain research --embed
# Query research
memphis ask "What did paper X say about Y?"
providers:
ollama:
url: http://localhost:11434/v1
model: qwen2.5:3b-instruct-q4_K_M
role: primary
memory:
path: ~/.memphis/chains
auto_git: false
embeddings:
backend: ollama
model: nomic-embed-text
multi_agent:
enabled: true
protocol: campfire-circle
self_loop:
enabled: true
auto_journal: true
โ
35+ CLI commands
โ
90.7% decision accuracy
โ
<200ms average response time
โ
Works with 1K-100K+ blocks
โ
8 chain types supported
โ
Multi-agent operational
โ
98.7% test coverage
Journal: 1,300+ blocks
Decisions: 100+ blocks
Ask: 100+ blocks
Share: 450+ blocks
Total: 2,000+ blocks (growing)
MIT License - use freely!
Created by: Memphis Team
Agents: Watra ๐ฅ + Memphis โณโฌกโ
Protocol: Campfire Circle ๐ฅ
Community: Oswobodzeni
# One command to install everything
clawhub install memphis
# Initialize
memphis init
# First memory
memphis journal "Hello Memphis!" --tags hello
# Done! ๐
Memphis - Your AI Brain, Locally ๐ฅ๐ง
Get started in 5 minutes!
Published by: Memphis Team
Version: 3.0.0
Status: PRODUCTION READY โ
Date: 2026-03-04